492 Harris: FACTORS INFLUENCING THE WEIGHT OF BEAN SEED 
order, ranging from +.005 to +.339 with the average value +.159. 
With two or three exceptions only, the correlation coefficients may 
be considered statistically significant in comparison with their 
probable errors. 
As a basis of comparison there are available from an earlier 
paper* the constants showing the relationship between the number 
of ovules and seeds per pod and the number of pods per plant in 
these same series. 
The results are exhibited graphically in Fics: 3 and 4. In 
these the circles on the ordinates joined by the heavy lines show 
on the scale to the left the values of the correlations for num- 
ber of pods per plant and seed weight arranged in the order of 
their magnitude. The circles connected by the narrow lines show 
on the same scale the magnitude of the correlations for ovules and 
seeds respectively. 
The bars indicating the averages show that the mean corre- 
lation for number of pods per plant and number of ovules per plant 
is numerically higher than that for number of pods per plant and 
seed weight. The mean correlation for number of pods per plant 
and seeds per pod is lower than that for pods per plant and seed 
weight. 
The magnitude of the differences is not, however, very large. 
The actual averages are:} 
For pods per plant and ovules per pod, 
7 = 1924 + .0124. 
For pods per plant and seeds per pod, 
r = .1327 + .o118. 
For pods per plant and seed weight, 
rf = .1592 + OIIQ. 
The difference between the mean value of the correlation for 
pods and ovules and pods and seed weight is therefore 
arris, J. Arthur. On the correlation between somatic characters and fer- 
tility. II. Illustrations from Phaseolus vulgaris. Amer. Jour. Bot. 1: 398-411. 1914 
t The mean values of the correlations for pods and ovules and pods and seeds 
given here differ slightly but insignificantly from those published elsewhere, because 
only 27 of the 32 series are used here, 
